Walkthroughs
Unscripted recordings of me building and testing automations in Glow.
4 recorded demos · Automation building and testing
Idea board (sebbuild.com) ↗My YouTube channel ↗Glow YouTube (@Human_Glow) ↗Claude Voice + MCP: Sticky Note to Self-Testing Automation in Glow
Loads YouTube only when you press play. Privacy
Reviewed English captions (SRT). Turn on CC in the player.
Written overview
Starting from a sticky note on the shared canvas, I operate Claude by voice through Glow MCP to build a Gmail and Glow AI classification automation, send real test emails through Gmail MCP, and add a Slack branch from a second sticky note.
What you will see
- Two live canvas sessions side by side, starting from a sticky note with Gmail labeling requirements
- Claude builds the automation through Glow MCP with schema validation and layout, then sends real test emails via Gmail MCP
- A second sticky note triggers a live voice redesign that branches urgent emails to Slack
- Claude Voice
- Glow MCP
- Multiplayer Canvas
- Gmail MCP Testing
Claude Voice + MCP: AI Lead Research Agent to Reusable Template
Loads YouTube only when you press play. Privacy
Reviewed English captions (SRT). Turn on CC in the player.
Written overview
Starting from a sticky note on the Glow canvas, I prompt Claude by voice over Glow MCP to build an inbound lead research automation with a native Glow AI Agent (Gemini Flash with web search), a personalized Gmail reply draft and a Slack summary DM. I test it end to end by asking Claude via Gmail MCP to send a realistic inquiry from Linear, then refactor the working automation into a reusable template with a Custom Variables step.
What you will see
- Claude reads the sticky note over Glow MCP and wires a Gmail trigger, a native Glow AI Agent with live web search, a Gmail reply draft and a Slack DM
- End-to-end test via Gmail MCP using a realistic Linear inquiry, verifying the web-researched Gmail draft and Slack high-fit summary on the first run
- Voice refactoring into a reusable template: switching from Live to Draft after the live-mutation guard and extracting hardcoded values into a Custom Variables step
- Claude Voice
- Glow MCP
- AI Lead Research
- Custom Variables Template
1-Prompt Inbound Deal Desk: Voice Builder & Streaming Architecture
Loads YouTube only when you press play. Privacy
English captions, auto-generated by YouTube. Use CC in the player; automatic captions may contain errors.
Written overview
I describe an inbound deal desk in a voice prompt. Glow builds the automation on the canvas, with parallel steps for lead scoring and budget extraction, Gmail drafts, Slack alerts and a test setup.
What you will see
- A spoken brief becomes an automation with configured steps
- Parallel branches: AI lead scoring and budget extraction alongside Gmail drafts and Slack alerts
- A test setup below the automation lets me try the generated steps
- Voice Builder
- Real-Time Streaming
- Parallel Execution
- Testing Simulator
Building Business Automation with Voice & Live Canvas MCP
Loads YouTube only when you press play. Privacy
English captions, auto-generated by YouTube. Use CC in the player; automatic captions may contain errors.
Written overview
A spoken brief becomes an automation on Glow's canvas. The recording shows customer order cleanup, Python scoring and an HTTP endpoint for the report.
What you will see
- Atomic step streaming and background step verification
- Deduplication, customer order cleanup, and Python scoring logic
- Branching logic and direct HTTP webhook deployment
- Voice Automation
- Atomic Canvas Streaming
- Python Step
- HTTP Webhook
Planned recordings
View 3 planned topics (not yet published)
Error handling that survives production
Retries, backoff, and knowing when to stop trying.
Connecting an API nobody documented
Reading live network responses and working backwards.
Running automation for many clients at once
Multi-tenant governance and keeping each client workspace sealed.
Let's talk about what you are building.
I'm not taking freelance work. Happy to talk about Glow, automation ideas, or a product engineering opportunity.
The hello is written. The rest is yours.
hozaksebastian@gmail.com
Webmail may ask you to sign in. Mail app requires a configured email application.